Corneal haze and neovascularization affect over 1.5 million Americans every year and are among the leading causes of blindness worldwide. Gene therapy is an attractive and novel approach to prevent/treat these corneal disorders. However, clinical utility of gene therapy is severely limited due to unavailability of tissue-targeted gene transfer methods. To develop tissue-targeted selective gene therapy approaches for the cornea, we hypothesized that foreign genes can be selectively expressed in the cornea at the desired site for selected duration using appropriate vectors and vector-delivery techniques. Preliminary in vivo experiments performed with AAV serotype 2 or 5, lentivirus or plasmid vectors containing CMV or hybrid CMV+chicken-3-actin promoter, and defined vector-delivery techniques demonstrated that transgene can be precisely expressed in keratocytes of normal/damaged corneas in vivo for the desired time period. Decorin (a small leucine-rich proteoglycan) gene therapy has been shown to prevent fibrosis and angiogenesis in various disease animal models. These reports led us to hypothesize that selective expression of decorin in keratocytes can inhibit/prevent corneal haze and neovascularization with minimal side effects. The in vitro studies performed to test this hypothesis demonstrated competence of decorin to inhibit keratocyte transformation to myofibroblasts. This transformation is known to cause corneal haze in vivo. We further hypothesize that selective tissue-targeted gene transfer approaches can be used to develop animal models for studying the specific function of disease-causing genes such as TGFp and BIGH3 in the adult cornea in vivo without altering their expression in vital organs. The function of such genes cannot be studied using conventional transgenic approaches because TGFp-deficient transgenic animals suffer lethal defects and die by 4 weeks of age. The specific aims to test the hypotheses are 1) vector and vector-delivery techniques regulate level, duration, and location of transgene expression in keratocytes in vivo 2) decorin gene therapy can control corneal haze and 3) decorin gene therapy can inhibit corneal neovascularization. Using a mouse model, tested vectors, and optimized vector delivery-techniques, we will test Specific Aim 1 and will define short- and long-term selective gene transfer approachesfor the cornea. Rabbit models will be used to test Specific Aims 2 and 3 by delivering decorin into keratocytes with optimal selective gene transfer methods.